Controversial

Entrapment
Neuropathies
WilliamW. Campbell, MD, COL, MC, USAa,
Mark E. Landau, MD, LTC, MC, USAb,*
KEYWORDS

There is no significant disagreement about the

major common entrapment neuropathies, such
as carpal tunnel syndrome (CTS), ulnar neuropathy
at the elbow, and peroneal neuropathy at the knee.
There is some minor disagreement about fine
points regarding the clinical presentation, electrodiagnosis, and management, but there is certainly
no major disagreement, such as whether these
entities exist. In contrast, there is a group of
entrapment syndromes about which there is major
disagreement, including whether or not they even
exist. The book entitled Tunnel Syndromes lists
some 25 tunnel syndromes, most of which are so
obscure that most neurologists have never heard
of them, for example, Alcocks tunnel syndrome.1
There are other entrapment syndromes about
which clinical questions arise on a regular basis,
and which are the subject of this discussion. These
include thoracic outlet syndrome (TOS), radial tunnel syndrome (RTS), ulnar nerve entrapment at the
arcade of Struthers, piriformis syndrome (PS), and
tarsal tunnel syndrome (TTS).

THORACIC OUTLET SYNDROME

TOS has a long and controversial history. It was
first described by Kinnier Wilson in approximately
1900, and there is some reason to think that one
of his original patients may have had CTS. TOS
was established in the minds of physicians as

a cause of acroparesthesias for some 40 odd

years before CTS and cervical radiculopathy
were described. What we now refer to as CTS
was referred to by various names in the early twentieth century, including tardy median palsy; it did
not receive its name until 1947 in a paper by Lord
Brain and colleagues.2 In patients who have
acroparesthesias, TOS is one of the first possibilities come to mind for many physicians, but for
most neurologists, it is nearly the last thing to
come to mind. The perception regarding the prevalence of TOS between surgeons and neurologists
is at variance on the order of 100:1. First rib resections and other surgical procedures continue to be
performed on a regular basis in the United States.
There are occasional disasters from these procedures, with severe damage to the brachial
plexus.3,4 It has been estimated that the likelihood
of postsurgical brachial plexopathy is about equal
to the likelihood of the diagnosis of TOS being
accurate in the first place. Cherington and Cherington5 found an interesting correlation between
TOS surgery and method of reimbursement in
Colorado, with Medicaid patients virtually never
receiving TOS surgical procedures.
Compression of the neurovascular structures at
the thoracic outlet has been blamed on various
structures, including cervical ribs or bands, an
elongated C7 transverse process, the scalenus
anterior muscle, the pectoralis minor tendon, and

a
Department of Neurology, Uniformed Services University of Health Sciences, 4301 Jones Bridge Road,
Bethesda, MD 20814, USA
b
Department of Neurology, Walter Reed Army Medical Center, 6900 Georgia Avenue, Washington, DC 20307,
USA
* Corresponding author.
E-mail address: mark.landau@amedd.army.mil (M.E. Landau).

Neurosurg Clin N Am 19 (2008) 597608

doi:10.1016/j.nec.2008.07.001
1042-3680/08/$ see front matter. Published by Elsevier Inc.

neurosurgery.theclinics.com

 Thoracic outlet syndrome  Radial tunnel syndrome

 Arcade of Struthers  Piriformis syndrome
 Tarsal tunnel syndrome

598

Campbell & Landau

the humeral head.6,7 We should probably recognize five types of TOS: arterial, venous, true neurogenic, disputed, and droopy shoulder syndrome.
Vascular TOS occurs most commonly in young
athletes, especially those who hyperabduct the
shoulder in throwing and in swimming.8 In the
position of hyperabduction, the humeral head
may compress the axillary artery or vein, causing
such vascular changes as stenosis, occlusion,
poststenotic dilatation, aneurysm formation, and
subclavian vein occlusion.9
True neurogenic TOS is an entity about which
there is substantial agreement. It is rare, attributable to compression of the lower trunk of the
plexus by a cervical rib or band. True neurogenic
TOS is more common in women and presents
with medial arm pain; medial hand and forearm
sensory dysfunction; and weakness and atrophy
of the thenar, hypothenar, and other intrinsic
hand muscles. The thenar eminence, muscles
supplied by the median nerve, are typically
involved earlier and more severely than the muscles of the hypothenar eminence. The electrodiagnostic abnormalities in true neurogenic TOS
include an absent or small medial antebrachial cutaneous amplitude (MABC), followed in decreasing
frequency by absent or small median motor, ulnar
sensory, and ulnar motor amplitudes.10 The MABC
nerve conduction study (NCS) abnormalities
closely parallel the median motor response
changes, leading some investigators to conclude
that the damage of neurogenic TOS is in a T1 myotomal distribution (Fig. 1).11 Needle electromyography (EMG) shows denervation and chronic
neurogenic motor unit potential changes in the distribution of the lower trunk of the plexus.10 One
typical pattern would include fibrillation potentials
and positive sharp waves in the abductor and opponens pollicis brevis muscles, associated with
more chronic denervative changes in the adductor
digiti minimi, the first dorsal interosseous, the
flexor pollicis longus, and, rarely, the extensor indicus proprius.
The proximal location for neurogenic TOS pathologic findings limits an electromyographers ability to detect subtle demyelinating changes.
Considering CTS, nerve conduction measurements are easily recorded for the sensory and
motor fibers of the median nerve as they transverse the wrist. This allows detection of mild demyelinating changes in the sensory fibers, with
excellent sensitivity and specificity for diagnosis.
Analogous studies for any proximal nerve
segments are invariably associated with technical
limitations. There is a long-standing myth that TOS
can cause slowing of the ulnar motor conduction
velocity when stimulating at Erbs point and doing

Fig. 1. Neurogenic TOS: ligamentous band from the

transverse process of C7 to the first thoracic rib (FTR)
lying in close approximation to the C8 and T1 anterior
primary rami and the proximal portion of the lower
trunk. (From Levin KH, Wilbourn AJ, Maggiano HJ. Cervical rib and median sternotomy-related brachial plexopathies: a reassessment. Neurology 1998;50:1412; with
permission.)

an across the plexus velocity determination. An

article was written describing this in the New
England Journal of Medicine in 1972.12 The article
was subsequently proved to be fraudulent
research with a fabricated figure and was withdrawn, with a stern editorial rebuke about
authorship responsibilities.13 This technique has
finally gradually fallen into disfavor and is rarely
performed currently by reputable electromyographers. Therefore, electrodiagnosis relies on the
detectable axonal changes distal to the site of
pathologic change resulting from Wallerian degeneration. This supports the argument for those who
favor TOS being relatively underdiagnosed in that
subtler disease is not detectable by means of
objective measures.14
Patients who neither have true vascular nor true
neurogenic TOS have been referred to as having
disputed neurogenic TOS. The leading surgical
proponents think that 97% of patients who have
TOS are neurologic, and disputed neurogenic
TOS accounts for most of these.14 These patients
are said to have aching pain radiating from the
scapular region down the arm; it can radiate
through the axillary region, the triceps area, and

Controversial Entrapment Neuropathies

the medial forearm. Arm flexion or abduction
worsens the pain. There may be paresthesias in
the medial forearm and hand. There are no motor
manifestations, and the results of electrodiagnostic studies are normal. In other words, the diagnosis relies on sensory signs and symptoms, which
are entirely subjective. One of the major criteria
for the diagnosis of TOS is reported to be a positive
Adson maneuver, a physical diagnostic sign that
was shown 50 years ago to be positive in 80% of
normal asymptomatic individuals. With the Adson
maneuver, the patients arm hangs at the side,
the head is rotated toward the affected side, and
the patient is instructed to breathe deeply. The
test is considered positive if the radial pulse disappears. In a presentation and abstract at the American Academy of Neurology (AAN) in 2005, some
investigators showed that positive TOS physical
test results can occur in patients who have CTS.15
Other provocative maneuvers purportedly abnormal in TOS include the Halstead maneuver, the
Wright test, and the Roos overhead exercise
test. A positive Halstead test result (exaggerated
military position or costoclavicular test) is defined
as a decreased radial pulse with the patient standing, with the shoulders far backward and in
a downward position so as to narrow the costoclavicular space. A positive Wright test result is also
defined as a decreased radial pulse, with the
shoulder hyperabducted and the elbow flexed.16
A positive Roos or elevated arm stress test
(EAST) is reproduction of the patients symptoms
following 3 minutes of rapid opening and closing
of the hand, with the arm abducted to 90 and
externally rotated and the elbow flexed to 90 .17
These test results may all be abnormal in healthy
subjects.18
Recently, neuroimaging has attempted to correlate anatomic variations with clinical symptoms.1921 Thoracic outlet MRI findings have
been compared in symptomatic and asymptomatic individuals. Demondion and colleagues20
found that patients who had TOS had a smaller
costoclavicular distance, a thicker subclavius
muscle, and a wider retropectoralis minor space
after hyperabduction and external rotation of the
arm compared with controls. Nerve compression
was noted in 7% of the symptomatic group and
in none of the asymptomatic group. The symptomatic group was defined has having two abnormal
provocative test results (Adson maneuver, Wright
maneuver, Eden maneuver, Roos test, or Tinel
sign in the supraclavicular fossa). Neurologic
examinations and electrodiagnostic studies were
not performed. In a similar study, Demirbag and
colleagues21 compared MRI findings of the
thoracic outlet in an adducted versus

hyperabducted position of patients and controls.

This study blinded an examiner in performing provocative tests. Interestingly, two thirds of subjects
in the control group had a positive Adson test
result. The specificity for the exaggerated military
position and hyperabduction test were considerably better, with 8.3% and 25%, respectively, of
controls having a positive test result. All subjects
had normal EMG findings, but no details were provided regarding the extent to which imitators, such
as CTS or cervical radiculopathy, were excluded.
The study concluded that considerably more
variation of predefined radiographic parameters
was seen in the MRI scans between the neutral
and hyperabducted position in the patient group.
Less variation of these findings was seen in the
control group. Furthermore, in the hyperabducted
position, compression of venous (52.3% versus
41.7%), arterial (23.0% versus 1.4%), and neural
(12.1% versus none) segments was more
common in the patient groups.
At this time, it is unclear whether a combination
of MRI findings and provocative clinical maneuvers can substantiate the diagnosis of disputed
neurogenic TOS. Surgical intervention is
frequently offered to patients who fail conservative
management, but the evidence for efficacy and the
perioperative complications vary considerably.
A surgeon who confidently supports the existence
of disputed TOS reports 90% success rates, with
perioperative complications less than 1%.14
Conversely, another study showed persistent
disability in 60% of patients 1 year after surgery
and perioperative complications greater than
30%.22 Complications reported in the literature
include postoperative brachial plexopathy, pleural
tears, pneumothorax, hemorrhage, and coronary
artery spasm.
The most basic of medical principles must be
invoked in any discussion of TOSfirst, do no
harm. Disputed TOS is a sensory syndrome without definitive objective findings to support its existence. The clinical tools used for its diagnosis are
exceptionally sensitive, with poor specificity. The
surgical treatment can result in objective deficits
in a not insignificant percentage of patients,
many of whom had a normal thoracic outlet before
surgery.

RADIAL TUNNEL SYNDROME

RTS is a condition in which there is chronic pain in
the lateral elbow region, allegedly because of
entrapment of elements of the radial nerve by
various structures in a tunnel in the region of the
supinator muscle.23 There is considerable doubt
as to whether this clinical entity or the anatomic

599

600

Campbell & Landau

tunnel exists.24,25 As with the other controversial
entrapments, it is essentially a pain syndrome
without objective clinical or electrophysiologic
manifestations. There is significant clinical overlap
between RTS and lateral epicondylitis (LE). LE, or
tennis elbow, is a common condition that results
from overuse of the extensor and supinator
muscles, typically by playing tennis. It produces
pain and tenderness over the region of the lateral
epicondyle and usually responds to conservative
treatment with rest and anti-inflammatory agents.
Examination of the elbow usually shows tenderness over the epicondyle, which is maximal just
distal to the epicondyle. With the elbow extended,
wrist extension or elbow supination against resistance causes increased pain in the symptomatic
area, and forceful wrist flexion or pronation
stretches the involved region and also causes
pain. Occasional patients fail to respond to
conservative therapy and develop chronic lateral
elbow pain sometimes referred to as resistant
tennis elbow. There is a school of orthopedic surgeons who believe that this syndrome of resistant
tennis elbow is attributable to entrapment of the
radial nerve in the radial tunnel.26,27
The radial tunnel is a nebulous anatomic passageway; nerve compression has been attributed
to no less than eight different anatomic structures
here. With anatomic studies, some groups found
that a radial tunnel existed in most subjects,
whereas other groups, seemingly equally competent, found that the radial tunnel existed in none
of their subject cadavers (Fig. 2). There is considerable doubt simply on an anatomic basis as to
whether there is such an entity as the radial tunnel.
The would-be tunnel is approximately 5 cm long
and located on the anterior aspect of the proximal
radius.28,29 It originates just proximal to the radial
capitulum and terminates distally at the point
where the posterior interosseous nerve (PIN)
passes deep to the superficial portion of the supinator muscle. The capsule of the radial capitulum
forms the floor of the tunnel. The brachioradialis,
extensor carpi radialis brevis (ECRB), and longus
muscles form the lateral wall, and the biceps tendon and brachialis muscle constitute the medial
wall. The anatomic structures implicated in PIN
entrapment include a fibrous band linking the brachioradialis to the brachialis, a cluster of blood
vessels overlying the nerve (leash of Henry), the
medial tendinous edge of the ECRB, the arcade
of Frohse, the distal border of the supinator muscle, fibrous bands within the supinator muscle,
and the septum between the extensor carpi ulnaris
and the extensor digitorum.
The chief symptom of RTS is pain over the
lateral aspect of the proximal forearm, with

Fig. 2. The radial nerve branches into the posterior

interosseous nerve and the superficial radial nerve.
The posterior interosseous nerve traverses deep to
the ligament of Frohse and the supinator muscle.
ECRL and ECRB, extensor carpi radialis longus and
brevis; EDC, extensor digitorum communicans. (From
Stanley J. Radial tunnel syndrome: a surgeons perspective. J Hand Ther 2006;19:181; with permission.)

occasional radiation distally or proximally. The

pain may be associated with a definitive injury or
after excessive use but not invariably. There are
none of the typical hallmarks of a PIN lesion,
such as finger extensor weakness (finger drop) to
a greater degree than wrist extensor weakness.
Proponents of RTS have suggested that the PIN
provides sensation for joints and muscles and
can produce pain when damaged, despite not providing cutaneous sensation. There are only subtle
differences between RTS and LE. In RTS, palpation tenderness is demonstrated at the point
where the radial nerve enters the supinator muscle, approximately 5 cm distal to the lateral epicondyle. In LE, the tenderness is directly over the
lateral epicondyle, the origin of the extensor muscles, as opposed to the muscles themselves.28 In
RTS, pain is elicited by fully extending the elbow,
followed by resistance against forearm supination.
Supposedly, contraction of the supinator increases compression of the arcade of Frohse on
the nerve.28 Resisting middle finger extension
with the forearm pronated and extended may
also elicit pain in the supinator region.

Controversial Entrapment Neuropathies

Electrodiagnostic testing is useful in excluding
other etiologies of upper extremity pain, because
there are no specific findings seen in RTS. Usually,
the results of the study are normal, without any
evidence of radial or PIN abnormalities. Some investigators have claimed slowing of radial motor
nerve conduction velocity with stimulation at the
spiral groove while recording over the extensor
digitorum communis.23
When conservative measures fail, some investigators recommend surgical exploration with radial tunnel release in an attempt to relieve the
symptoms.30,31 The same procedure can also relieve the symptoms of LE, because division of
the superficial portion of the supinator muscle
also relieves tension on the lateral epicondyle.
Posterior interosseous neuropathy may occur as
a complication of radial tunnel release.

Ligament of Struthers and Arcade of Struthers

Nerve Injuries
The ligament of Struthers and the arcade of
Struthers are two different anatomic structures
that are often confused. The ligament of Struthers
is a fibrous band that extends from an anomalous
supracondylar spur on the humerus down to the
medial epicondyle, which may compress the median nerve (Fig. 3). Struthers32 described this ligament and has been recognized eponymically for it.
Approximately 1% of the population has this
anomaly, but only rarely does it cause nerve injury.33 In contrast, the arcade of Struthers consists
of filmy, occasionally substantive, tissue along the
medial intermuscular septum, which may compress the ulnar nerve (Fig. 4). It is in no way related
to the ligament of Struthers. In fact, Struthers did
not describe the tissue that later came to be called
the arcade of Struthers.34 The eponymic designation is erroneous.
As the ulnar nerve descends through the upper
arm, it moves from the anterior to the posterior
compartment through the intermuscular septum.
The so-called arcade of Struthers is an aponeurotic strand that extends from the medial intermuscular septum to the medial head of the triceps,
about 8 cm proximal to the medial epicondyle. In
one anatomic study, it was seen in 13.5% of the
limbs, but in no instance did it compress the ulnar
nerve.35
Proximal median mononeuropathy
by the ligament of Struthers
The median nerve, along with the brachial artery,
runs beneath the ligament of Struthers, but only
rarely does it cause median nerve injury. Plain radiographs reveal the supracondylar spur, but the
ligament may be present without a spur.

Fig. 3. Ligament of Struthers between the supracondylar spur (*) and medial epicondyle (m). Median nerve
(n) and brachial artery (ba). (From Aydinlioglu A.
Bilateral median nerve compression at the level of
Struthers ligament. Case report. J Neurosurg 2000;
92:694; with permission.)

Symptoms include pain above the elbow and local

tenderness in addition to neurologic deficits in the
distribution of the median nerve. There is not much
controversy regarding this entity when there is
objective evidence of median neuropathy.
Ulnar mononeuropathy attributable
to the arcade of Struthers
During ulnar transposition surgery, the medial intermuscular septum must be released or the ulnar
nerve may become secondarily compressed after
it is transposed anteriorly. In such cases re-exploration of the nerve shows deformity and compression of the nerve at the point of entrapment by the
inadequately released medial intermuscular septum, and this has been referred to as the arcade
of Struthers. This posttransposition syndrome is
well recognized and not a subject for debate.3640
The debate concerns whether or not the medial
intermuscular septum or an arcade of Struthers
can cause a spontaneous compression syndrome
of the ulnar nerve.35 There is disagreement in the
literature about the existence of the arcade of
Struthers in a patient who has not had an ulnar
transposition. Some anatomic studies have found
tissue consistent with an arcade of Struthers in anatomic specimens, whereas other investigators
have reported that it virtually never exists.4144
The situation is analogous to that with RTS. It is

601

602

Campbell & Landau

Fig. 4. Arcade of Struthers is located
between the medial head of the triceps
and the medial intermuscular septum.
The ulnar nerve lies deep. (From Kane E,
Kaplan EB, Spinner M. Sur le trajet du
nerf cubital au niveau du bras [Observations on the course of the ulnar nerve in
the arm]. Ann Chir 1973;27:48796; with
permission.)

unclear anatomically whether the structure exists.

Spinner45 suggested that a friction neuritis could
develop from repeated movement of the nerve
against bony irregularities of the distal humerus
and that an arcade of Struthers could be a contributing factor. The arcade of Struthers, as a mechanism of ulnar compression in the unoperated arm,
is extraordinarily rare or a myth, with some other
etiology more likely responsible. The first author
of the only three cases reported in peer-reviewed
literature is Ochiai.46,47 The clinical findings were
identical to ulnar mononeuropathy at the elbow,
but the electrodiagnostic findings showed slowing
of nerve conduction velocity or conduction block
70 to 100 mm proximal to the medial epicondyle.

PIRIFORMIS SYNDROME
The PS is a condition allegedly attributable to compression of the sciatic nerve by the piriformis muscle, causing pain in the buttock radiating down the
posterior lower extremity, simulating radiculopathy. There is some problem with terminology.
The term piriformis syndrome should only be

used to designate proximal sciatic neuropathies

in which the piriformis muscle itself is involved in
the etiology. There are many other causes of proximal sciatic neuropathy, such as gluteal trauma
with hematoma, injection palsy, compression,
hip arthroplasty, hip fracture, or endometrial
implant.48 Although the sciatic nerve may be involved near the piriformis in these conditions,
none should be legitimately referred to as PS.
The piriformis is one of an array of small muscles
that lie deep in the gluteal region: the pectineus,
piriformis, obturators, gemelli, and quadratus
femoris. These short muscles are probably more
important as postural stability muscles for the hip
joint rather than as prime movers. The piriformis
originates from the anterior sacral surface, passes
through the greater sciatic foramen, and inserts
onto the medial side of the upper border of the
greater trochanter (Fig. 5). The piriformis rotates
the extended thigh laterally but acts as an abductor when the hip is flexed. Just below the piriformis
lies the gemellus superior. At its exit from the
pelvis, the sciatic nerve exists as two discrete fascicles: the peroneal division and the tibial division.

Controversial Entrapment Neuropathies

Fig. 5. Piriformis muscle overlying the sciatic nerve.

(From Benzon HT. Piriformis Syndrome: Anatomic concsiderations, a New Injection Technique, and a Review
of the Literature. Anesthesiology 2003;98:1443; with
permission.)

These are most often encased in a common

epineural sheath and appear to be a single nerve.
The two separate components can easily be
dissected apart, however. Anomalies of the exit
pattern of the sciatic nerve from the pelvis can
occur.49 The normal pattern is for the intact sciatic
nerve to exit just below the piriformis muscle in the
groove between the piriformis and the gemellus
superior. In some patients, the sciatic nerve splits:
one division exits below the piriformis muscle, and
the other exits above it. In some patients, the piriformis is also bifid and one or both elements of the
sciatic nerve pierce the muscle. The entire intact
nerve may exit directly through the middle of the
belly of the piriformis, or one division may exit
through the belly of muscle, whereas the other
division exits above or below. The incidence of
this anomaly varies widely by ethnic origin. In a series of American cadavers, there was some anomaly of sciatic nerve exit in 10%.50 The piriformis is
muscle is a small structure, perhaps 3 to 5 cm wide
at its widest point and perhaps 1 cm thick. The sciatic nerve at that point is a large robust structure.
There are numerous problems with the concept
of the PS, and a great debate on how common this
entity is.51 It is not uncommon to see articles in the
literature that stress the uncertainty about the syndrome (eg, The piriformis syndrome is overdiagnosed versus The piriformis syndrome is
underdiagnosed or The piriformis syndrome
myth or reality?)5254 Most estimates of the incidence of PS in patients with back pain estimate
its incidence at 1% to 2% of cases, but some
practitioners claim to see dozens to hundreds of
patients per year. One problem with PS is that
the nerve is large in relation to the muscle, and it
is difficult to see how such a small and relatively
insubstantial structure as the piriformis muscle

could significantly compress the sciatic nerve.

The PS was first described in 1928, long before
lumbosacral radiculopathy was recognized as
a clinical entity, and long before radiologic imaging
or electrodiagnosis existed. Discogenic pain,
arising from the annulus itself, without a radicular
component, can radiate as far as the knee, causing radiating pain in the buttock and leg but without objective evidence of radiculopathy. Pain in
the buttock and leg when sitting is one of the classic symptoms of PS but is also quite common in
radiculopathy. Many, perhaps most, patients
who have lumbosacral radiculopathy attributable
to disc herniation have maximal pain in the sitting
position.
The pathophysiologic mechanism of sciatic
involvement in PS is unclear. Some investigators
believe the presence of an anatomic anomaly is
irrelevant,53 although others contend that PS
requires some exit anomaly.55 Nobody has
described an abnormal leg movement, such as
the external rotation that would accompany a contraction of the piriformis, as part of the syndrome.
Therefore the spasm must be isometric, without
muscle shortening, and it is therefore difficult to
envision how nerve compression could occur. It
would make just as much sense to claim that
spasm of the gemellus superior pressing up
against the nerve could cause the nerve compression symptoms. The PS is virtually the only compression or entrapment neuropathy attributable
to a muscle. Virtually every other one is attributable
to compression by a fascial band, aponeurosis,
ligament, or other such rigid structure (eg, transverse carpal ligament, humeroulnar arcade,
arcade of Frohse, sublimis bridge, lancinate ligament). Nevertheless, the proponents of PS believe
that one of the largest and most robust nerves in
the body is being compressed by a soft and compliant structure, which is extremely atypical in the
class of disorders that we ordinarily include as
entrapment neuropathies.
There are two clinical scenarios that need to be
considered. One is the patient with physical examination and electrodiagnostic evidence of proximal
sciatic neuropathy. Can the mechanism be PS in
the absence of other etiologies of nerve damage?
The other clinical scenario regards patients who
have buttock pain or buttock pain with radiation
down the posterior leg; no other typical signs or
symptoms of nerve injury, including paresthesias,
hypesthesias, hyporeflexia, weakness or atrophy;
and a normal electrodiagnostic assessment.
What is the likelihood that a chronic or chronic
recurrent compression of the sciatic nerve could
account for the pain without the other neurologic
manifestations? There is little definitive evidence

603

604

Campbell & Landau

in the literature to support the diagnosis of PS in
either case. Even its proponents refer to it as
a loose cluster of symptoms.56 The clinical criteria include pain in the buttock and usually some
part of the course of the sciatic nerve distal to it,
tenderness in the region of the intersection of the
piriformis muscle and the sciatic nerve, and positive straight leg raise at 15 less than on the unaffected side, or less than 60 when PS is bilateral.
None of these have been critically evaluated, and
are certainly not specific. The piriformis muscle is
small and lies deep beneath the massive gluteus
maximus; yet, proponents claim to be able to palpate piriformis spasm in patients who have the
syndrome (the princess and the pea sign).
Some of the physical maneuvers allegedly indicative of the syndrome are contradictory. The Pace
test consists of sciatic pain produced by abduction of the hip against resistance, presumably
causing the muscle to contract against the nerve,
reproducing the symptoms.57 Yet another maneuver, the Freiburg test, reproduces the pain with
forced internal rotation, stretching the muscle.58
It would seem the Freiburg test should, if anything,
decompress the nerve by stretching the muscle,
removing the pressure, and improving the symptoms. How these two diametrically opposite
maneuvers could reproduce the pain is difficult
to reconcile. In the Beatty maneuver, the patient
lies painful side up with the leg flexed and the hip
abducted, holding the knee on the painful side
several inches off the table; this allegedly reproduces the pain.59 There have been no validation
studies of any of these tests.
Electrophysiologic criteria have been proposed
to assist in diagnosis of PS.56,60 In the flexed,
adducted, and internally rotated (FAIR) test, the
posterior tibial H reflex is measured with the patient prone and legs in the neutral position, and
then again with the hip flexed, adducted, and internally rotated. A latency prolongation of greater
than 3 SDs is stated to have greater than 80% sensitivity and specificity. There are multiple limitations, however, to performing H reflexes. One is
that they are not routinely obtainable in patients
older than the age of 60 years.61 Second, a prolongation of latency is not specific for the location of
the abnormality, because lesions at the nerve
root or plexus could also cause prolongations.
Whether or not lesions at these locations can
also become apparent only in the postures of the
FAIR test has not been systematically evaluated.
One needs to question the validity of the test as
performed by the proponents of the test. The
mean age of the patients in their two studies was
older than 50 years, yet they did not describe
one case in which the H reflex could not be

found.56,60 Furthermore, they claimed an ability

to attain H waves in greater than 300 subjects for
the peroneal nerve; yet, it is universally accepted
by electrodiagnosticians that peroneal H waves
are rarely obtainable. Finally, the waveforms
depicted in their studies were biphasic with a negative initial deflection, whereas most H waves are
triphasic with an initial positive deflection.
Treatments recommended for PS include physical therapy, local injections, botulinum toxin injections, and surgery. We are somewhat placated by
the 94% of 918 patients referred to physicians who
strongly believe in this entity, getting conservative
therapy only, without surgical interventions.60

TARSAL TUNNEL SYNDROME

The TTS is attributable to compression of the tibial
nerve within a fibro-osseous passageway behind
the medial malleolus. There is widespread agreement regarding the existence of the syndrome,
but there are differences of opinion regarding its
epidemiology as an etiology for foot pain and
paresthesias, particularly in diabetics.
The roof of the tarsal tunnel is the flexor retinaculum (lancinate ligament) and the tendinous arch
of the abductor hallucis (AH). The lancinate ligament is thin and insubstantial compared with the
transverse carpal ligament. The posterior tibial
nerve passes under the ligament and then divides
into larger medial and smaller lateral plantar
branches. In 7% of individuals, the tibial nerve
branches before the tarsal tunnel.62 The medial
plantar branch innervates the AH, flexor digitorum
brevis, flexor hallucis brevis, and the first lumbrical, and it provides sensation to the medial part
of the sole of the foot and the medial three toes.
The lateral plantar branch innervates the rest of
the deep muscles of the foot and provides sensation to the lateral part of the sole of the foot and the
lateral two toes. The calcaneal branch provides
sensation to the heel; it usually arises proximal to
the tarsal tunnel, but the origin is variable.
The common symptoms of TTS are pain in the
area of the medial malleolus with occasional radiation into the medial aspect of the foot. To be convinced of nerve involvement, there should be
paresthesias or dysesthesias on the plantar
aspect of the foot. Although intrinsic foot muscles
may become weak or atrophic, these manifestations rarely interfere with function and go overlooked by patients. The objective physical
findings in TTS are few. Weakness of the foot muscles cannot be demonstrated on neurologic examination. There may be hypesthesia over the sole,
especially in the medial plantar distribution, usually
with sparing of the heel (calcaneal innervation).

Controversial Entrapment Neuropathies

There may be tenderness behind the medial
malleolus and a Tinels sign over the tarsal tunnel.
A stress test similar to the Phalen test for CTS has
been described. The ankle is passively maximally
everted and dorsiflexed, the toes are pulled up,
and this position is held for 5 to 10 seconds to
reproduce the symptoms.63
The differential diagnosis includes arthritis, plantar fasciitis, plantar callosities, vascular disease, or
small fiber sensory neuropathies. The diagnosis is
often made to explain foot pain of no other apparent origin, even in absence of any neurologic
deficit. The causes of TTS include external compression by ill-fitting footwear or casts, ankle
trauma, thickened flexor retinaculum, ganglia,
posttraumatic fibrosis, and others.64 The plantar
nerves may be injured distal to the tarsal tunnel
from local trauma, compression by tendon sheath
cysts, or AH hypertrophy.64
There are electrodiagnostic findings that can
support the clinical diagnosis of TTS. Ideally, the
most sensitive test would detect conduction slowing of sensory nerve fibers as they traverse the tarsal tunnel. This, however, is rarely seen. In most
cases of TTS, the medial and lateral plantar sensory
nerve action potential (SNAPs) are absent.65 This
abnormality is not specific for TTS because it may
be a manifestation of sensory polyneuropathies or
even seen in asymptomatic individuals, particularly
those older than the age of 50 years. Prolongation
of the medial and lateral plantar motor latencies
has been noted in approximately 20% to 50% of
individuals with TTS.65,66 No published electrodiagnostic studies include needle EMG in the evaluation of TTS.67 There are many in the
electrodiagnostic community who believe that findings of denervation in the intrinsic foot muscles are
common in asymptomatic individuals; therefore,
any needle electrode findings would be of little
use.68,69 Others, however, have disputed this
claim.70 A pattern of EMG abnormalities that would
support the diagnosis includes denervation of intrinsic foot muscles isolated to the symptomatic
limb of patients who have unilateral disease, with
sparing of abnormalities of the extensor digitorum
brevis (EDB). The EDB is the only nontibial-innervated intrinsic foot muscle. The electrodiagnostic
tests are particularly useful for ruling out other neurologic etiologies of foot pain, such as sensory polyneuropathies and radiculopathies.
When conservative measures fail to limit symptomatology, surgical intervention may be necessary. A review of published articles showed
symptomatic improvement in 91% of patients
surgically treated.71 This is a commonly performed
procedure, especially by podiatrists and orthopedic foot surgeons.72

Surgical decompression of the tarsal tunnel in

isolation or along with other peripheral nerves
has been used as an alternative approach to
diabetic neuropathy.7378 The hypothesis that
underlies this approach is that patients who have
polyneuropathy may have increased susceptibility
to compression mononeuropathies at potential
sites of entrapment. Thus, peripheral nerve
surgery for diabetics has also included decompression and neurolysis of the common peroneal
nerve at the fibular head and the deep peroneal
nerve in the anterior tarsal region.73 Note that the
sural nerve, which provides sensation to the lateral
foot, and the saphenous nerve, which provides
sensation to the medial lower leg, were not
included. It is not clear how surgery could help
sensation or other parameters in these regions of
diabetics. Dellon and his coworkers7478 have
been the major proponents of this intervention.
He and others claim that nerve decompressions
in diabetics improve proprioception, cutaneous
sensation, balance, and neuropathic pain and prevent ulcerations.
The AAN issued a practice advisory regarding
decompression surgery for the treatment of
diabetic neuropathy.79,80 Per this systematic
review, the current evidence supporting the utility
of decompressive surgery for the treatment of diabetic neuropathy is of poor quality and design. All
available studies for this intervention were categorized as class IV studies, and the AAN concluded
that this treatment alternative be considered unproved. There was one study that was originally
categorized as class III because it was prospective
with a blinded evaluator.77 It was ultimately downgraded to class IV because it lacked detail regarding electrodiagnostic studies and lacked clarity
segregating patients into polyneuropathy versus
compression categories. Additionally, scales
used to define improvement were deemed
arbitrary. In this study, 79% of surgically decompressed nerves showed improvement in the twopoint discrimination test. All other studies were
nonblind case series using a variety of outcome
measures. The panel recommended prospective
randomized controlled trials with standardized
outcome measures to determine the value of this
intervention.

REFERENCES
1. Pecina MM, Krmpotic-Nemanic J, Markiewitz AD.
Tunnel syndromesperipheral nerve compression
syndromes. 3rd edition. Boca Raton (FL): CRC Press
LLC; 2001.

605

606

Campbell & Landau

2. Brain WR, Wright AD, Wilkinson M. Spontaneous
compression of both median nerves in the carpal
tunnel. Lancet 1947;1:27782.
3. Wilbourn AJ. Thoracic outlet syndrome surgery
causing severe brachial plexopathy. Muscle Nerve
1988;11:6674.
4. Cherington M, Happer I, Machanic B, et al. Surgery
for thoracic outlet syndrome may be hazardous to
your health. Muscle Nerve 1986;9:6324.
5. Cherington M, Cherington C. Thoracic outlet syndrome: reimbursement patterns and patient profiles.
Neurology 1992;42:9435.
6. Liu JE, Tahmoush AJ, Roos DB, et al. Shoulder-arm
pain from cervical bands and scalene muscle anomalies. J Neurol Sci 1995;128:17580.
7. Korkmaz N, Ozcakar L. Long cervical costae articulating with the first ribs: a salient case of thoracic outlet syndrome. Am J Phys Med Rehabil 2006;85(1):
104.
8. Katirji B, Hardy RW Jr. Classic neurogenic thoracic
outlet syndrome in a competitive swimmer: a true
scalenus anticus syndrome. Muscle Nerve 1995;
18:22933.
9. Rigatelli G, Rinuncini M, Roncon L, et al. Images in
cardiovascular medicine. Functional subclavian
artery compression caused by thoracic outlet
syndrome. Circulation 2005;112(17):e2801.
10. Aminoff MJ, Olney RK, Parry GJ, et al. Relative utility
of different electrophysiologic techniques in the
evaluation of brachial plexopathies. Neurology
1988;38:54650.
11. Levin KH, Wilbourn AJ, Maggiano HJ. Cervical rib
and median sternotomy-related brachial plexopathies: a reassessment. Neurology 1998;50(5):
140713.
12. Urschel HC Jr, Razzuk MA. Management of the thoracic-outlet syndrome. N Engl J Med 1972;286:
11403.
13. Wilbourn AJ, Lederman RJ. Evidence for conduction
delay in thoracic-outlet syndrome is challenged. [letter]. N Engl J Med 1984;310:10523.
14. Roos DB. Thoracic outlet syndrome is underdiagnosed. Muscle Nerve 1999;22:1269 [discussion
1367].
15. Nord KM, Kothari MJ, Kapoor P, et al. Determination
of false-positive rate of thoracic outlet syndrome
diagnostic maneuvers in carpal tunnel syndrome
patients. [abstract]. Neurology 2005;64:A287.
16. Wright IS. The neurovascular syndrome produced
by hyperabduction of the arms. Am Heart J 1945;
29:119.
17. Roos DB. New concepts of TOS that explain etiology, symptoms, diagnosis and treatment. Vasc
Surg 1979;13:31321.
18. Plewa MC, Delinger M. The false-positive rate of
thoracic outlet syndrome shoulder maneuvers in
healthy subjects. Acad Emerg Med 1998;5:33742.

19. Remy-Jardin M, Remy J, Masson P, et al. Helical CT

angiography of thoracic outlet syndrome. AJR Am
J Roentgenol 2000;174:166774.
20. Demondion X, Bacqueville E, Paul C, et al. Thoracic outlet: assessment with MR imaging in asymptomatic and symptomatic populations. Radiology
2003;227:4618.
21. Demirbag D, Unlu E, Ozdemir F, et al. The relationship between magnetic resonance imaging findings
and postural maneuver and physical examination
tests in patients with thoracic outlet syndrome: results of a double-blind, controlled study. Arch Phys
Med Rehabil 2007;88:84451.
22. Franklin GM, Fulton-Kehoe D, Bradley C, et al. Outcome of surgery for thoracic outlet syndrome in
Washington state workers compensation. Neurology 2000;54:12527.
23. Roles NC, Maudsley RH. Radial tunnel syndrome:
resistant tennis elbow as a nerve entrapment.
J Bone Joint Surg Br 1972;54(3):499508.
24. van Rossum J, Buruma OJ, Kamphuisen HA, et al.
Tennis elbowa radial tunnel syndrome? J Bone
Joint Surg Br 1978;60-B(2):1978.
25. Rosenbaum R. Disputed radial tunnel syndrome.
Muscle Nerve 1999;22(7):9607.
26. Moss SH, Switzer HE. Radial tunnel syndrome:
a spectrum of clinical presentations. J Hand Surg
[Am] 1983;8(4):41420.
27. Sarhadi NS, Korday SN, Bainbridge LC. Radial tunnel syndrome: diagnosis and management. J Hand
Surg [Br] 1998;23(5):6179.
28. Dawson DM, Hallett M, Wilbourn AJ. Entrapment
neuropathies. 3rd edition. Philadelphia: LippincottRaven Publishers; 1999. p. 2149.
29. Lister GD, Belsole RB, Kleinert HE. The radial tunnel
syndrome. J Hand Surg [Am] 1979;4:529.
30. Sotereanos DG, Varitimidis SE, Giannakopoulos PN,
et al. Results of surgical treatment for radial tunnel
syndrome. J Hand Surg [Am] 1999;24:56670.
31. Jebson PJ, Engber WD. Radial tunnel syndrome:
long-term results of surgical decompression.
J Hand Surg [Am] 1997;22:88996.
32. Struthers J. On a peculiarity of the humerus and
humeral artery. Mon J Med Sci 1848;28:2647.
33. Terry RJ. A study of the supracondyloid process in
the living. Am J Phys Anthropol 1921;4:12936.
34. De Jesus R, Dellon AL. Historic origin of the
Arcade of Struthers. J Hand Surg [Am] 2003;
28(3):52831.
35. Siqueira MG, Martins RS. The controversial arcade
of Struthers. Surg Neurol 2005;64(Suppl 1):
S1721.
36. Rogers MR, Bergfield TG, Aulicino PL. The failed
ulnar nerve transposition. Etiology and treatment.
Clin Orthop Relat Res 1991;269:193200.
37. Kleinman WB. Revision ulnar neuroplasty. Hand Clin
1994;10:46177.

Controversial Entrapment Neuropathies

38. Vogel RB, Nossaman BC, Rayan GM. Revision anterior submuscular transposition of the ulnar nerve for
failed subcutaneous transposition. Br J Plast Surg
2004;57:3116.
39. Broudy AS, Leffert RD, Smith RJ. Technical problems
with ulnar nerve transposition at the elbow: findings
and results of reoperation. J Hand Surg [Am] 1978;
3:859.
40. Caputo AE, Watson HK. Subcutaneous anterior
transposition of the ulnar nerve for failed decompression of cubital tunnel syndrome. J Hand Surg
[Am] 2000;25:54451.
41. al Qattan MM, Murray KA. The arcade of Struthers: an
anatomical study. J Hand Surg [Br] 1991;16(3):3114.
42. Bartels RH, Grotenhuis JA, Kauer JM. The arcade of
Struthers: an anatomical study. Acta Neurochir
(Wien) 2003;145(4):295300.
43. von Schroeder HP, Scheker LR. Redefining the
Arcade of Struthers. J Hand Surg [Am] 2003;
28(6):101821.
44. Wehrli L, Oberlin C. The internal brachial ligament
versus the arcade of Struthers: an anatomical study.
Plast Reconstr Surg 2005;115(2):4717.
45. Spinner M. Injuries to the major branches of
peripheral nerves of the forearm. 2nd edition.
Philadelphia: W.B. Saunders; 1978. p. 2347.
46. Ochiai N, Hayashi T, Ninomiya S. High ulnar nerve
palsy caused by the arcade of Struthers. J Hand
Surg 1992;17B:62931.
47. Ochiai N, Honmo J, Tsujino A, et al. Electrodiagnosis
in entrapment neuropathy by the arcade of
Struthers. Clin Orthop Relat Res 2000;378:12935.
48. Yuen EC, So YT. Sciatic neuropathy. Neurol Clin
1999;17:61731.
49. Beaton LE, Anson BJ. The relationship of the sciatic
nerve and of its subdivisions to the piriformis muscle. Anat Rec 1937;70:13.
50. Beaton LE, Anson BJ. The sciatic nerve and the piriformis muscle: Their interrelation a possible cause of
coccygodynia. J Bone Joint Surg Am 1938;20:6868.
51. Silver JK, Leadbetter WB. Piriformis syndrome:
assessment of current practice and literature review.
Orthopedics 1998;21(10):11335.
52. Stewart JD. The piriformis syndrome is overdiagnosed. Muscle Nerve 2003;28(5):6446.
53. Fishman LM, Schaefer MP. The piriformis syndrome is
underdiagnosed. Muscle Nerve 2003;28(5):6469.
54. McCrory P. The piriformis syndromemyth or reality? Br J Sports Med 2001;35(4):20910.
55. Pecina M. Contribution to the etiological explanation
of the piriformis syndrome. Acta Anat (Basel) 1979;
105(2):1817.
56. Fishman LM, Zybert PA. Electrophysiologic
evidence of piriformis syndrome. Arch Phys Med
Rehabil 1992;73(4):35964.
57. Pace J, Nagle D. Piriformis syndrome. West J Med
1976;124:4359.

58. Freiberg AH. Sciatic pain and its relief by operation

on muscle and fascia. Arch Surg 1937;34:33749.
59. Beatty RA. The piriformis muscle syndrome: a simple
diagnostic maneuver. Neurosurgery 1994;34(3):
5124.
60. Fishman LM, Dombi GW, Michaelsen C, et al. Piriformis syndrome: diagnosis, treatment, and outcome
a 10 year study. Arch Phys Med Rehabil 2003;83:
295301.
61. DeLisa JA, Hang JL, Baran EM, et al. Manual of
nerve conduction velocity and clinical neurophysiology. 3rd edition. Philadelphia: Lippincott Williams
& Wilkins; 1994. p. 16879.
62. Havel PE, Ebraheim NA, Clark SE, et al. Tibial nerve
branching in the tarsal tunnel. Foot Ankle 1988;9:
1179.
63. Reade BM, Longo DC, Keller MC. Tarsal tunnel
syndrome. Clin Podiatr Med Surg 2001;18(3):
395408.
64. Stewart JD. Focal peripheral neuropathies. 1st
edition. New York: Elsevier; 1987. p. 3134.
65. Kinoshita M, Okuda R, Morikawa J, et al. The dorsiflexion-eversion test for diagnosis of tarsal tunnel
syndrome. J Bone Joint Surg Am 2001;83-A(12):
18359.
66. Galardi G, Amadio S, Maderna L, et al. Electrophysiologic studies in tarsal tunnel syndrome. Diagnostic
reliability of motor distal latency, mixed nerve and
sensory nerve conduction studies. Am J Phys Med
Rehabil 1994;73:1938.
67. Oh SJ, Sarala PK, Kuba T, et al. Tarsal tunnel syndrome: electrophysiological study. Ann Neurol
1979;5(4):32730.
68. Patel AT, Gaines K, Malamut R, et al. Usefulness of
electrodiagnostic techniques in the evaluation of
suspected tarsal tunnel syndrome: an evidencebased review. Muscle Nerve 2005;32:23640.
69. Falck B, Alaranta H. Fibrillation potentials, positive
sharp waves and fasciculation in the intrinsic muscles of the foot in healthy subjects. J Neurol Neurosurg Psychiatry 1983;46:6813.
70. Gatens PF, Saeed MA. Electromyographic findings
in the intrinsic muscles of normal feet. Arch Phys
Med Rehabil 1982;63:3178.
71. Dumitru D, Diaz CA, King JC. Prevalence of denervation in paraspinal and foot intrinsic musculature.
Am J Phys Med Rehabil 2001;80:48290.
72. Cimino WR. Tarsal tunnel syndrome: review of the
literature. Foot Ankle 1990;11:4752.
73. Bailie DS, Kelikian AS. Tarsal tunnel syndrome: diagnosis, surgical technique, and functional outcome.
Foot Ankle Int 1998;19(2):6572.
74. Wood WA, Wood MA. Decompression of peripheral
nerves for diabetic neuropathy in the lower extremity.
J foot Ankle Surg 2003;42:26875.
75. Dellon AL. Neurosurgical prevention of ulceration
and amputation by decompression of lower

607

608

Campbell & Landau

extremity peripheral nerves in diabetic neuropathy:
update 2006. Acta neurochir Suppl 2007;100:
14951.
76. Ducic I, Taylor NS, Dellon AL. Relationship between
peripheral nerve decompression and gain of pedal
sensibility and balance in patients with peripheral
neuropathy. Ann Plast Surg 2006;56:14550.
77. Caffee HH. Treatment of diabetic neuropathy by
decompression of the posterior tibial nerve. Plast
Reconstr Surg 2000;106:8135.
78. Aszmann OC, Kress KM, Dellon AL. Results of
decompression of peripheral nerves in diabetics:

a prospective blinded study. Plast Reconstr Surg

2000;106:81622.
79. Dellon AL. Treatment of symptomatic diabetic neuropathy by surgical decompression of multiple peripheral nerves. Plast Reconstr Surg 1992;89:
68997.
80. Chaudhry V, Stevens JC, Kincaid J, et al. Practice
advisory: utility of surgical decompression for
treatment of diabetic neuropathy. Report of the
Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 2006;66:18058.